Impact of climate change on a landslide in South East France, simulated using different GCM scenarios and downscaling methods for local precipitation

摘要

ABSTRACT: The aim of this paper is to assess the influence of different climate scenarios on scenarios for the impact variable ?landslide activity1. For this purpose, a site-specific model was used, relating the activity of a landslide in South East France to climate. Landslide activity was reconstructed from tree ring data. Hydrological field data indicated that the controlling climatic variable is net precipitation (precipitation minus evapotranspiration). However, this variable and hence the impact model could not explain all of the variations in landslide activity. The landslide model was fed with 1 temperature and several precipitation scenarios obtained by applying 3 different methods for downscaling 3 different general circulation model (GCM) simulations of the large-scale climate. The skill of the downscaling methods in reproducing the historical local precipitation was either limited or trivial, but fair enough to justify further application. The resulting scenarios for landslide activity were quite similar, with the exception of 2 specific combinations of GCM and downscaling method. Furthermore, short-term climatic variation, plausibly represented in one of the downscaling methods as a random noise component, caused additional variation in the resulting scenarios. The amount of variation in the climate scenarios is of the same order of magnitude as that in the landslide model. The general conclusion is not to focus on calibrating impact models while using only 1 climate scenario, but to assess the overall uncertainty of the impact scenario by considering different parameter settings of the impact model as well as different climate scenarios, as was done in the present study.